Qwen 2.5 Math 14B Iter 2
Collection
Qwen 2.5 is missing it's 14B and 32B math variants!! I have taken it upon myself to create them :) These are the Iteration 2 Models
•
4 items
•
Updated
•
1
Uploaded model
- Developed by: qingy2019
- License: apache-2.0
- Finetuned from model : unsloth/qwen2.5-14b-bnb-4bit
Huge thanks to Unsloth and the Huggingface TRL library.
This model is Qwen 2.5 14B fine tuned for a full epoch on the high quality garage-bAInd/Open-Platypus dataset for STEM reasoning.
| Training Detail | Value |
|---|---|
| Epochs | 1 |
| Steps | 2077 |
| Loss | 0.4218 |
| Batch size | 4 |
| Gradient Acc. Steps | 3 |
| Learning Rate | 2e-4 |
| LR Scheduler | cosine |
| Rank | 32 |
| Rank-Stabilized LoRA | Yes |
| Warm up steps | 5 |
| Weight Decay | 0.01 |
| Seed | 3407 |
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 35.46 |
| IFEval (0-Shot) | 59.81 |
| BBH (3-Shot) | 47.75 |
| MATH Lvl 5 (4-Shot) | 23.11 |
| GPQA (0-shot) | 16.00 |
| MuSR (0-shot) | 17.95 |
| MMLU-PRO (5-shot) | 48.12 |
- Downloads last month
- 55
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for qingy2024/Qwen2.5-Math-14B-Instruct-Alpha
Collection including qingy2024/Qwen2.5-Math-14B-Instruct-Alpha
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard59.810
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard47.750
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard23.110
- acc_norm on GPQA (0-shot)Open LLM Leaderboard16.000
- acc_norm on MuSR (0-shot)Open LLM Leaderboard17.950
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard48.120